AutoML Leaderboard

Best model	name	model_type	metric_type	metric_value	train_time
	1_Baseline	Baseline	auc	0.5	1.17
	2_DecisionTree	Decision Tree	auc	0.739429	9.5
	3_Linear	Linear	auc	0.86439	15.09
	4_Default_Xgboost	Xgboost	auc	0.867461	287.44
	5_Default_NeuralNetwork	Neural Network	auc	0.869571	9.99
	6_Default_RandomForest	Random Forest	auc	0.794363	24.86
the best	Ensemble	Ensemble	auc	0.88475	0.55

AutoML Performance

AutoML Performance Boxplot

Features Importance

Spearman Correlation of Models

Summary of 5_Default_NeuralNetwork

<< Go back

Neural Network

• n_jobs: -1
• dense_1_size: 32
• dense_2_size: 16
• learning_rate: 0.05
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

9.1 seconds

Metric details

	score	threshold
logloss	0.498088	nan
auc	0.869571	nan
f1	0.794915	0.61198
accuracy	0.84	0.6751
precision	1	1
recall	1	4.70116e-09
mcc	0.681178	0.6751

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.498088	nan
auc	0.869571	nan
f1	0.793103	0.6751
accuracy	0.84	0.6751
precision	0.910891	0.6751
recall	0.70229	0.6751
mcc	0.681178	0.6751

Confusion matrix (at threshold=0.6751)

	Predicted as 0	Predicted as 1
Labeled as 0	800	45
Labeled as 1	195	460

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 1_Baseline

<< Go back

Baseline Classifier (Baseline)

• n_jobs: -1
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

0.5 seconds

Metric details

	score	threshold
logloss	0.685103	nan
auc	0.5	nan
f1	0.607889	0.3928
accuracy	0.436667	0.3928
precision	0.436667	0.3928
recall	1	0.3928
mcc	0	0.3928

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.685103	nan
auc	0.5	nan
f1	0.607889	0.3928
accuracy	0.436667	0.3928
precision	0.436667	0.3928
recall	1	0.3928
mcc	0	0.3928

Confusion matrix (at threshold=0.3928)

	Predicted as 0	Predicted as 1
Labeled as 0	0	845
Labeled as 1	0	655

Learning curves

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of Ensemble

<< Go back

Ensemble structure

Model	Weight
3_Linear	1
4_Default_Xgboost	1
5_Default_NeuralNetwork	1

Metric details

	score	threshold
logloss	0.395402	nan
auc	0.88475	nan
f1	0.805071	0.439696
accuracy	0.842	0.519909
precision	1	0.973635
recall	1	0.00300079
mcc	0.685303	0.610451

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.395402	nan
auc	0.88475	nan
f1	0.802005	0.519909
accuracy	0.842	0.519909
precision	0.885609	0.519909
recall	0.732824	0.519909
mcc	0.680847	0.519909

Confusion matrix (at threshold=0.519909)

	Predicted as 0	Predicted as 1
Labeled as 0	783	62
Labeled as 1	175	480

Learning curves

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 2_DecisionTree

<< Go back

Decision Tree

• n_jobs: -1
• criterion: gini
• max_depth: 3
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

8.6 seconds

Metric details

	score	threshold
logloss	0.583441	nan
auc	0.739429	nan
f1	0.656958	0.260268
accuracy	0.704667	0.581966
precision	0.746512	0.581966
recall	1	0.105973
mcc	0.396031	0.581966

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.583441	nan
auc	0.739429	nan
f1	0.591705	0.581966
accuracy	0.704667	0.581966
precision	0.746512	0.581966
recall	0.490076	0.581966
mcc	0.396031	0.581966

Confusion matrix (at threshold=0.581966)

	Predicted as 0	Predicted as 1
Labeled as 0	736	109
Labeled as 1	334	321

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 6_Default_RandomForest

<< Go back

Random Forest

• n_jobs: -1
• criterion: gini
• max_features: 0.9
• min_samples_split: 30
• max_depth: 4
• eval_metric_name: auc
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

24.0 seconds

Metric details

	score	threshold
logloss	0.53895	nan
auc	0.794363	nan
f1	0.695218	0.396989
accuracy	0.742667	0.59434
precision	0.958824	0.800951
recall	1	0.0745929
mcc	0.501343	0.626023

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.53895	nan
auc	0.794363	nan
f1	0.613226	0.59434
accuracy	0.742667	0.59434
precision	0.892128	0.59434
recall	0.467176	0.59434
mcc	0.500005	0.59434

Confusion matrix (at threshold=0.59434)

	Predicted as 0	Predicted as 1
Labeled as 0	808	37
Labeled as 1	349	306

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 4_Default_Xgboost

<< Go back

Extreme Gradient Boosting (Xgboost)

• n_jobs: -1
• objective: binary:logistic
• eta: 0.075
• max_depth: 6
• min_child_weight: 1
• subsample: 1.0
• colsample_bytree: 1.0
• eval_metric: auc
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

286.5 seconds

Metric details

	score	threshold
logloss	0.429753	nan
auc	0.867461	nan
f1	0.780811	0.505597
accuracy	0.826667	0.562002
precision	1	0.931576
recall	1	0.00658826
mcc	0.654215	0.58899

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.429753	nan
auc	0.867461	nan
f1	0.775087	0.562002
accuracy	0.826667	0.562002
precision	0.894212	0.562002
recall	0.683969	0.562002
mcc	0.653299	0.562002

Confusion matrix (at threshold=0.562002)

	Predicted as 0	Predicted as 1
Labeled as 0	792	53
Labeled as 1	207	448

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back

Summary of 3_Linear

<< Go back

Logistic Regression (Linear)

• n_jobs: -1
• explain_level: 2

Validation

• validation_type: split
• train_ratio: 0.75
• shuffle: True
• stratify: True

Optimized metric

auc

Training time

14.2 seconds

Metric details

	score	threshold
logloss	0.493363	nan
auc	0.86439	nan
f1	0.775849	0.450781
accuracy	0.816667	0.645845
precision	0.969512	0.981788
recall	1	9.51433e-05
mcc	0.627684	0.645845

Metric details with threshold from accuracy metric

	score	threshold
logloss	0.493363	nan
auc	0.86439	nan
f1	0.770259	0.645845
accuracy	0.816667	0.645845
precision	0.850554	0.645845
recall	0.703817	0.645845
mcc	0.627684	0.645845

Confusion matrix (at threshold=0.645845)

	Predicted as 0	Predicted as 1
Labeled as 0	764	81
Labeled as 1	194	461

Learning curves

Permutation-based Importance

Confusion Matrix

Normalized Confusion Matrix

ROC Curve

Kolmogorov-Smirnov Statistic

Precision-Recall Curve

Calibration Curve

Cumulative Gains Curve

Lift Curve

<< Go back